(GRS 414) Proceedings of the Workshop on Integral Experiment Covariance Data for Critical Safety Validation, 9 - 11 March 2016, GRS Garching
For some time, attempts to quantify the statistical dependencies of critical experiments and to account for them properly in validation procedures were discussed in the literature by various groups. Besides the development of suitable methods especially the quality and modeling issues of the freely available experimental data are in the focus of current discussions, carried out for example in the Expert Group on Uncertainty Analysis for Criticality Safety Assessment (UACSA) of the OECD-NEA Nuclear Science Committee. The same committee compiles and publishes also the freely available experimental data in the International Handbook of Evaluated Criticality Safety Benchmark Experiments. Most of these experiments were performed as series and might share parts of experimental setups leading to correlated results. The quality of the determination of these correlations and the underlying covariance data depend strongly on the quality of the documentation of experiments.